DCEG has a long and distinguished history of investigating the relationship between treatments administered to patients to control an initial cancer, and the risk of subsequently developing a second cancer. This represents a unique observational situation in which one can study the consequences of human exposure to well defined chemical carcinogens and to ionizing radiation. The opportunity now exists to move these studies into the genetic arena, by investigating the relationship between common polymorphisms in genes affecting the bioavailability of chemical carcinogens and various outcomes of clinical interest. Such studies could identify population sub-groups which are at particular risk of second cancers, thrombotic events, acute myelosuppression, response to treatment or even survival. The potential also exists to identify genetic variants which may reduce the risk of adverse outcomes. Information of this kind could have a significant impact on clinical decision-making. (a) CGB's first pharmacogenetics project is now well underway. We have begun a study of genetic polymorphisms in the genes related to tamoxifen and estrogen bioavailability and the risk of developing endometrial cancer and breast cancer as a result of exposure to tamoxifen. Clinical trials have demonstrated that women exposed to tamoxifen are at increased risk of developing endometrial cancer and at decreased risk of developing breast cancer. In general terms, these differences have been attributed to tissue specific variations in whether tamoxifen acts as an estrogen agonist or an estrogen antagonist. We hypothesize that genetic variations in the genes which affect tamoxifen and/or estrogen metabolism may identify sub-groups of women who are more or less likely to benefit from the administration of tamoxifen. The first phase of this project is targeting the women who participated in NSABP's Tamoxifen Breast Cancer Prevention Trial ("P1"), with invasive breast cancer the first outcome to be investigated. A "case-only" study design has been chosen. This proposal is now undergoing IRB review at NSABP. We have been granted access to DNA specimens that were previously extracted and anonymized for another study. The development of the assays for the 30 polymorphisms selected for study is complete, and they are now being validated using CEPH families. Genotyping will begin as soon as IRB approval has been obtained. Subsequent phases of this project may include: (a) analysis of the non-invasive breast cancers which developed in P1 subjects; (b) analysis of the endometrial neoplasms from this same study; and (c) genotyping of a random sample of the entire P1 cohort, to permit seeking evidence of a major effect for candidate genes, and to obtain the absolute risk estimates required (if evidence of a gene-tamoxifen interaction is found) to apply these findings clinically. (b) Another variation on the theme of common variants within less penetrant genes as modifiers of cancer risk is our ongoing study of genetic polymorphisms in genes which are part of the IGF1 signaling pathway. Elevated levels of IGF1, a cytokine with both mitogenic and anti-apoptotic effects, have been associated with increased risks of a variety of different cancers, including premenopausal breast, colon, prostate and lung cancer. In collaboration with the Core Genotyping Facility and the Office of the Director, Epidemiology & Biostatistics Program, genetic variations in these genes are being systematically identified and then studied as determinants of neoplasm risk among participants of the Prostate, Lung, Colon and Ovarian Cancer (PLCO)screening trial. The first analysis will target PLCO participants who were found to have either colorectal cancer or advanced adenomatous polyps at the time of their baseline colonscopy. IGF1/IGF-BP3 levels are also being studied among participants in the Honolulu Heart Cohort who developed prostate cancer during prospective follow-up. (c) The next major pharmacogenetic study now under development will target treatment-related myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Genes involved in the activation and detoxification of specific chemotherapeutic agents, as well as DNA repair genes, will be evaluated as modifiers of the risk of MDS/AML in this setting. Case material will be drawn from the data bases maintained by the national cooperative clinical trial groups. NSABP is the first group to join this project; others are expected to follow. This project has been developed by Dr. Charles Rabkin, during his sabbatical year with CGB. (d) It is anticipated that opportunities for other analyses of the relationship between common genetic variants and the risk of various cancer treatment-related outcomes will be sought and identified. Particularly promising is the opportunity of working with various national clinical trials cooperative groups, both to explore the possibilities of using archived tumor samples as a source of DNA for gene study, as well as to consider the prospective collection of germline DNA from participants in selected clinical trials. Among the ideas under consideration are an evaluation of genetic influences upon the risk of developing contralateral breast cancer among women with breast cancer participating in clinical trials, the option of extending our studies of tamoxifen/gene interactions relative to the endometrial neoplasms which developed in P1 trial participants, and the possibility of studying genetic determinants of bleomycin-related pulmonary fibrosis among men treated for testicular cancer. (e) CGB staff have provided cancer genetics consultation to the Food and Drug Administration in support of its investigation of a cluster of malignant lymphomas which has been identified among patients exposed to a novel class of therapeutic agents, the tumor necrosis factor inhibitors. These medications have been approved recently for the treatment of refractory rheumatoid arthritis and for severe Crohn's disease. Thirty-nine persons with lymphoma have been identified by MedWatch, FDA's post-market adverse event surveillance system. A manuscript is now in press describing the findings of the initial investigation. Available data do not permit drawing a firm conclusion regarding the etiologic relationship between exposure to these medications and the subsequent development of lymphoma, but there is sufficient cause for concern to warrant additional analytic epidemiologic study in an effort to resolve this important question. (f) An excess of Ewing's sarcoma as a second primary malignancy following retinoblastoma has been reported by a CGB staff member. An excess risk of Merkel cell carcinoma, a rare cutaneous neuroendocrine malignancy, was observed in patients who were immunologically compromised as a result of infection with HIV.